The three common ways to buy software engineering services from a partner look interchangeable on a rate card and behave very differently in practice. Picking the wrong one costs you 20–40% on the same engineering output. Here's how we help clients choose.
TL;DR: Use staff augmentation when scope is fluid and your in-house team owns delivery. Use a dedicated team when you need a self-contained pod that owns a product area. Use fixed-price only when scope is genuinely locked — which is rare.
The three models, plainly
Staff augmentation
You get one or more individual engineers slotted into your existing team, reporting to your tech leads, using your Jira and your repo. You pay a monthly rate per engineer. The partner is a staffing conduit; delivery responsibility stays with you.
- Typical duration: 3–12 months per engineer.
- Typical cost: fixed monthly rate per head; easy to predict.
- Who owns quality: your in-house leads (partner provides the engineer but not the delivery plan).
Dedicated team
You get a pod — say, 2 engineers + 1 QA + a part-time delivery manager — operating as a self-contained unit. The pod owns a product area end-to-end, attends your standups, but has its own internal rituals. You pay a monthly blended rate.
- Typical duration: 6–24 months.
- Typical cost: fixed monthly; includes DM overhead and QA.
- Who owns quality: shared — partner DM owns delivery, you own product direction.
Fixed-price
You agree a scope, price and deadline up front. The partner delivers or eats the overrun. Changes cost extra (change requests).
- Typical duration: 4–16 weeks per milestone.
- Typical cost: locked at signing.
- Who owns quality: the partner, against the acceptance criteria.
The decision matrix
Five dimensions matter. Score your project honestly on each.
| Dimension | Staff aug wins | Dedicated team wins | Fixed-price wins |
|---|---|---|---|
| Scope clarity | Low-medium | Medium | High |
| In-house engineering capacity | Enough to lead | Thin | Product-thin |
| Duration | Any | 6+ months | Short (< 4 months) |
| Change tolerance | High | Medium | Low |
| Budget predictability needed | Medium | Medium | High |
When each model actually wins
Staff augmentation wins when:
- You have a strong tech lead or CTO but missing bandwidth at the IC level.
- Scope is roadmap-driven and changes sprint-to-sprint.
- You care about having your team learn and own the code, not just receive it.
Dedicated team wins when:
- You're a non-technical founder or a small team without a senior engineer to supervise ICs.
- You have a product area (mobile, platform, AI) that deserves its own pod.
- You want the partner to own delivery rituals, not just ship code.
Fixed-price wins when:
- Scope is genuinely locked — e.g., a WordPress migration, a specific integration, a defined MVP module.
- You need absolute budget certainty for board or finance reasons.
- The work is well-understood (not research-heavy or AI-R&D).
Hidden cost behaviours
Staff augmentation appears cheapest per head. The hidden cost is your own management time — budget 20% of a senior's week for review and mentoring. If you don't have that senior, augmentation ends in pain.
Dedicated teams come with DM overhead priced in. The hidden cost is ramp — the first 4 weeks deliver about 60% of steady-state velocity. After that, throughput often beats augmentation on a per-£ basis.
Fixed-price looks safest. The hidden cost is change-request friction — scope drift triggers renegotiation, which kills trust. Fixed-price works well only when scope truly is fixed, which is rare in product work.
Our honest recommendation pattern
After 150+ engagements across all three models, here's what we usually suggest:
- Seed / pre-seed: fixed-price MVP milestone first, graduating to dedicated team after product-market signal.
- Series A: dedicated team for a product area + staff aug for in-house capacity gaps.
- Series B+: staff aug to flex in-house team; dedicated pods for bets that need isolation (AI R&D, new platform).
- Enterprise: usually starts fixed-price for the first workstream, moves to dedicated pods as trust builds.
FAQ
Can we switch models mid-engagement?
Yes, and clients often do. A common path is fixed-price MVP → 3-month dedicated pod → 12-month dedicated pod + 1 staff-aug engineer.
Which model is best for an MVP?
Fixed-price if scope is clear; dedicated team if you want the partner to shape scope with you. Never staff-aug an MVP to a vendor because you lack the internal leadership to run one yourself.
How does Krapton handle scope change in fixed-price?
Clear change-request process: written, priced, signed. No verbal add-ons. Keeps scope drift from eroding the relationship.
Next step
We'll walk your project through the decision matrix in a 30-minute call and recommend the model we think fits — including saying "you don't need us, hire in-house" when that's true. Book a consultation, explore our services catalogue, or browse the bench.
